How A Two-Stroke Expansion Chamber Works

The Five Sections Of An Expansion Chamber

A two-stroke has no exhaust valve. The piston uncovers the exhaust port on the way down and covers it again on the way up, and for a stretch of every revolution the exhaust port and the transfer ports are open at the same time. Without help, some of the fresh fuel, air and oil coming up the transfers would simply flow straight out the exhaust. The expansion chamber is that help. It is shaped so the exhaust pulse bounces back off it at the right moments.

Every tuned pipe has the same five parts, in this order from the cylinder:

1. Header. A nearly straight tube, usually diverging between 0 and 2 degrees, that carries the first violent pulse away from the port.
2. Diffuser. A long, widening cone, often built from two or three rolled cones welded together. Modern pipes diverge at roughly 7 to 9 degrees. Gordon Jennings used 8 degrees in his 1964 Cycle World method and ended the cone at about 6 times the header area.
3. Belly. The fat middle. It sets the distance between the two cones.
4. Baffle cone. A shorter, steeper narrowing cone, about 16 to 25 degrees in modern practice. Jennings used 15 degrees.
5. Stinger. A small outlet tube that bleeds gas into the silencer. A common rule puts its diameter at 0.59 to 0.63 times the header diameter and its length at about 12 times its own diameter. Jennings gave 1.25 inches as an empirical outlet size for a 250.

The silencer bolts on after the stinger. On a two-stroke it mostly handles noise; the chamber does the tuning. Our silencer repack guide covers that last part.

Watch The Waves

Press the steps in order. Red is a pressure wave, blue is a suction wave, and amber is fresh charge. The strip under the pipe is one crank revolution from top dead center to top dead center, with the exhaust and transfer windows marked.

1. Exhaust opens. The piston edge uncovers the port while cylinder pressure is still high. The gas leaves as a strong pressure pulse that runs down the header.
2. Diffuser echo. As the pipe widens, part of that pulse reflects back toward the cylinder as suction. It reaches the port while the transfers are open and helps pull the burnt gas out and the fresh charge up. Some fresh charge follows the burnt gas into the header.
3. Baffle cone echo. Where the pipe narrows, part of the pulse reflects back as pressure. If it arrives after the transfers close but before the exhaust port shuts, it pushes the spilled charge back into the cylinder and holds it there.
4. Port closes. The piston covers the exhaust port with more charge trapped than the engine could have managed with a plain pipe.

Wikipedia's two-stroke article puts the suction from a racing pipe at about 7 psi below atmosphere near bottom dead center, with the transfers nearly wide open. That pull is a big part of why a two-stroke "comes on the pipe" so hard.

Why A Cone Sends A Wave Back

A pressure wave travelling down a tube partly reflects every time the cross-section changes. The rule Jennings explained in 1964 is simple. A section that gets wider behaves a little like an open pipe end and returns a suction wave. A section that gets narrower behaves a little like a closed end and returns a pressure wave. A long, gentle cone spreads the reflection over time, which is why a slow diffuser gives a wider powerband than a short, steep one.

One detail surprises people. The gas itself does not race through the pipe each cycle. The wave does. According to the Wikipedia summary of chamber design, the gas leaving the stinger was made two or three cycles earlier.

Pipe Length And Tuned Rpm

A wave takes a fixed time to run to a cone and back. The engine does not take a fixed time to go around. That is the whole story of pipe tuning. Jennings' method has two steps:

1. Time for a crank angle: t = (degrees / 360) / (rpm / 60).
2. Distance to the reflecting section: L = (wave speed × t) / 2, halved because the wave goes down and back.

Jennings used about 1,700 ft/s (518 m/s) for the wave speed in most of the pipe and about 3,500 ft/s for the first foot of very hot gas near the port. In his worked example for a Villiers 250 peaking at 6,500 rpm, the transfers closed 140 degrees after the exhaust opened. That is about 0.0035 seconds, so the baffle cone reflection had to come from about 35.7 inches down the pipe.

Our example pipe in the figures uses one average wave speed of 518 m/s to keep the math visible. With the exhaust opening at 85 degrees after top dead center and the transfers closing at 242.5 degrees, the middle of our baffle cone sits 827.5 mm from the port, so the pressure wave lands right as the transfers close at about 8,200 rpm. The suction echo from the middle of the diffuser, 478 mm from the port, lands near bottom dead center at the same speed.

Real pipe designers go much further. Gordon P. Blair's Design and Simulation of Two-Stroke Engines (SAE R-161) models the unsteady gas flow through ports and pipes with simulation instead of a single wave speed, and accounts for temperature, bends and multi-step cones. The simple method still explains the behavior you feel on the bike.

The Rpm Figure: When Does The Echo Arrive?

Drag the slider. The red marker is the moment the baffle cone's pressure wave gets back to the port. At low rpm the crank has not turned far when it arrives, so it shows up early, while the transfers are still feeding the cylinder. At high rpm the crank has gone further around, so the wave arrives after the port has already shut and the spilled charge is lost.

In our example the pressure wave lands inside the useful window from roughly 7,000 to just under 10,000 rpm. Below and above that, the pipe stops helping. That band is the powerband. A longer pipe moves it down the rev range, and a shorter pipe moves it up. Temperature moves it too, because the speed of sound in a gas rises with the square root of its absolute temperature, so a hotter pipe behaves like a shorter one.

What Hurts A Pipe

• Dents. Any change in diameter reflects part of the wave. A crushed belly or diffuser adds a reflection in the wrong place and weakens the ones you want. KTM's troubleshooting table for the 2024 250 and 300 SX and XC lists a leaky or deformed exhaust system as a cause of low engine power.
• Leaks. A cracked weld or a header joint that is not sealing bleeds off pressure. Look for black soot or oily streaks at joints. A crankcase pressure test with a block-off plate on the exhaust side will also show a header leak.
• A blocked spark arrester or wrong stinger. Too much restriction at the outlet raises chamber pressure and heat. Highland Cycles had a 2024 300 XC-W at 89 hours that suddenly ran flat and quiet. The screen-style spark arrester was caked shut, and burning it clean brought most of the power back (Highland Cycles, YouTube, at 1:11).
• A pipe tuned for something else. A pipe built for another engine or rpm range puts the waves in the wrong place in the cycle.

Honda's 2025 owner's manual for the CRF250R motocross race bike inspects the exhaust pipe and muffler every race, which Honda puts at about 2.5 hours. Check the pipe after any crash or rock strike, too.

Expansion Chamber Myths

• "A pipe works by back pressure." The main effect is wave timing. A well designed chamber's average pressure rise is much smaller than a muffler's.
• "A bigger pipe makes more power everywhere." A pipe moves power to the rpm its lengths are tuned for. It takes from somewhere else to give there.
• "Small dents don't matter." Every area change reflects some wave energy. How much power a given dent costs depends on its size and place, and we have not found a source that puts a number on it.

Some engines also change the pipe's behavior with rpm. Honda's ATAC and Suzuki's SAEC opened extra chamber volume at low rpm. Most modern dirt bikes do the job at the port instead, with a power valve.

Track Exhaust Service By Hours

The pipe itself has no service interval, but the jobs around it do. On KTM's 2024 250 SX motocross bike, the silencer filling is changed every 45 hours and the piston every 15. KTM's 2026 300 XC-W enduro bike goes 90 and 45 (PDF p. 55). MotoMind logs engine hours by hand or from its GPS ride tracker and shows each task as upcoming, due or overdue, so you can see the next repack and top end coming.

FAQ

What does an expansion chamber do on a two-stroke?

It uses the exhaust pulse as a pressure wave. The widening diffuser sends a suction wave back that helps draw fresh charge into the cylinder, and the narrowing baffle cone sends a pressure wave back that pushes escaped charge into the cylinder just before the exhaust port closes.

How do you calculate two-stroke pipe length?

Jennings’ method: find the time for the crank angle you want, t = (degrees / 360) / (rpm / 60), then multiply by the wave speed and halve it, L = (speed × t) / 2. He used about 1,700 ft/s for the wave speed. Simulation methods such as Blair’s refine this.

Does a dent in a two-stroke pipe matter?

Yes. Any change in diameter reflects part of the pressure wave, so a dent changes the timing and strength of the echoes. KTM lists a deformed exhaust system as a cause of low engine power.

Why does a two-stroke come on the pipe?

At the rpm the pipe is tuned for, the reflected waves arrive at the port at the right moments, so the cylinder fills better and holds more charge. Below that rpm the echoes come back early, above it too late.

Is a two-stroke pipe the same as a muffler?

No. The expansion chamber tunes the engine with pressure waves. The silencer after the stinger mainly reduces noise and needs its packing replaced on a schedule.

Does a hotter pipe change the powerband?

Yes. Wave speed rises with gas temperature, roughly with the square root of absolute temperature, so a hotter pipe acts like a shorter one and tunes for higher rpm.